Commits

Branch selector

 master

User selector

rryan

Datepicker

All time

Commit History

Commits on Oct 5, 2017

• Save an unnecessary logical_not in the maximum/minimum gradient.

  Show description for 376147c
  

  rryan

  authored and

  tensorflower-gardener

  committed
  376147c

  Copy full SHA for 376147c

• Update the tf.contrib.signal guide to include guidance on computing Mel spectrograms and MFCCs.

  Show description for f6b15b0
  

  rryan

  authored and

  tensorflower-gardener

  committed
  f6b15b0

  Copy full SHA for f6b15b0

Commits on Oct 4, 2017

• [tf-signal] Use tf.spectral.dct in mfccs_from_log_mel_spectrograms instead of a private implementation.

  Show description for add6d2d
  

  rryan

  authored and

  tensorflower-gardener

  committed
  add6d2d

  Copy full SHA for add6d2d

Commits on Oct 3, 2017

• Add tf.spectral.dct, based on scipy.fftpack.dct.

  Show description for cb460e4
  

  rryan

  authored and

  tensorflower-gardener

  committed
  cb460e4

  Copy full SHA for cb460e4

• [tf-signal] Add a test that windowing, framing, and mel ops are constant foldable for constant inputs.

  Show description for 991dea6
  

  rryan

  authored and

  tensorflower-gardener

  committed
  991dea6

  Copy full SHA for 991dea6

Commits on Oct 2, 2017

• [tf-signal] Add tf.contrib.signal.mfccs_from_log_mel_spectrograms.

  Show description for f94d410
  

  rryan

  authored and

  tensorflower-gardener

  committed
  f94d410

  Copy full SHA for f94d410

• [tf-signal] Remove checks that frame_length <= fft_length in stft and inverse_stft.

  Show description for 09d0c5f
  

  rryan

  authored and

  tensorflower-gardener

  committed
  09d0c5f

  Copy full SHA for 09d0c5f

Commits on Sep 26, 2017

• Fix incorrect tensor_util.constant_value results for Pack ops with axis != 0.

  Show description for 044a117
  

  rryan

  authored and

  tensorflower-gardener

  committed
  044a117

  Copy full SHA for 044a117

Commits on Sep 13, 2017

• Add mel-scale conversion matrix support to tf.contrib.signal.

  Show description for a4f6e7c
  

  rryan

  authored and

  tensorflower-gardener

  committed
  a4f6e7c

  Copy full SHA for a4f6e7c

Commits on Aug 9, 2017

• Remove newlines from tf.nn.dynamic_rnn args list.

  Show description for 26e628b
  

  rryan

  authored and

  tensorflower-gardener

  committed
  26e628b

  Copy full SHA for 26e628b

Commits on Aug 1, 2017

• Fix some typos in StreamExecutor's cuFFT support.

  Show description for 5202a5b
  

  rryan

  authored and

  tensorflower-gardener

  committed
  5202a5b

  Copy full SHA for 5202a5b

Commits on Jul 29, 2017

• Add a tf.contrib.signal API guide.

  Show description for f8b575d
  

  rryan

  authored and

  tensorflower-gardener

  committed
  f8b575d

  Copy full SHA for f8b575d

• Remove "small" test size for window_ops_test.

  Show description for 25ed0f2
  

  rryan

  authored and

  tensorflower-gardener

  committed
  25ed0f2

  Copy full SHA for 25ed0f2

Commits on Jul 28, 2017

• Increase bound on tf.contrib.signal.inverse_stft gradient error to avoid flakiness on macOS.

  Show description for d629a83
  

  rryan

  authored and

  tensorflower-gardener

  committed
  d629a83

  Copy full SHA for d629a83

Commits on Jul 25, 2017

• Disable padding by default in tf.contrib.signal.stft to match tf.contrib.signal.frame.

  Show description for 18158c8
  

  rryan

  authored and

  Amit Patankar

  committed
  18158c8

  Copy full SHA for 18158c8

• Add Short-time Fourier Transform (STFT) and inverse STFT support to tf.contrib.signal.

  Show description for 8b8ee45
  

  rryan

  authored and

  Amit Patankar

  committed
  8b8ee45

  Copy full SHA for 8b8ee45

• Speed up tf.contrib.signal spectral_ops_test.py by reducing the size of the gradient test.

  Show description for 879e207
  

  rryan

  authored and

  tensorflower-gardener

  committed
  879e207

  Copy full SHA for 879e207

• Reduce gather_op_test timeouts by reducing the size of testHigherRank.

  Show description for abcc81d
  

  rryan

  authored and

  tensorflower-gardener

  committed
  abcc81d

  Copy full SHA for abcc81d

• Update documentation of various dimension/axis arguments to mention support for negative axis indexing.

  Show description for 7108dce
  

  rryan

  authored and

  tensorflower-gardener

  committed
  7108dce

  Copy full SHA for 7108dce

Commits on Jul 22, 2017

• Add a note about tf.contrib.signal to the 1.3 release notes.

  rryan

  committed
  643fe1f

  Copy full SHA for 643fe1f

Commits on Jul 20, 2017

• Documentation fixes and polishing for tf.contrib.signal.

  Show description for 97cf16b
  

  rryan

  authored and

  tensorflower-gardener

  committed
  97cf16b

  Copy full SHA for 97cf16b

Commits on Jul 19, 2017

• Disable padding by default in tf.contrib.signal.stft to match tf.contrib.signal.frame.

  Show description for 03a790d
  

  rryan

  authored and

  tensorflower-gardener

  committed
  03a790d

  Copy full SHA for 03a790d

• Add Short-time Fourier Transform (STFT) and inverse STFT support to tf.contrib.signal.

  Show description for d70b39b
  

  rryan

  authored and

  tensorflower-gardener

  committed
  d70b39b

  Copy full SHA for d70b39b

Commits on Jul 14, 2017

• Add overlap_and_add to tf.contrib.signal.

  Show description for ff5b2bf
  

  rryan

  authored and

  tensorflower-gardener

  committed
  ff5b2bf

  Copy full SHA for ff5b2bf

• Add new features to tf.contrib.signal.frames:

  Show description for 5051e97
  

  rryan

  authored and

  tensorflower-gardener

  committed
  5051e97

  Copy full SHA for 5051e97

• Add support for generating Hann and Hamming windows to tf.contrib.signal.

  Show description for 3aa588a
  

  rryan

  authored and

  tensorflower-gardener

  committed
  3aa588a

  Copy full SHA for 3aa588a

Commits on Jul 11, 2017

• Add an axis parameter to tf.gather. Fixes GitHub issue #11223.

  Show description for b1f9e2c
  

  rryan

  authored and

  tensorflower-gardener

  committed
  b1f9e2c

  Copy full SHA for b1f9e2c

Commits on Jul 10, 2017

• Add a PadV2 op with support for specifying a pad value.

  Show description for a6773e9
  

  rryan

  authored and

  tensorflower-gardener

  committed
  a6773e9

  Copy full SHA for a6773e9

Commits on Jun 12, 2017

• Fix incorrect results from Eigen-based IRFFT kernel when input dimensions exceed fft_length.

  Show description for 1bb1c0d
  

  rryan

  authored and

  tensorflower-gardener

  committed
  1bb1c0d

  Copy full SHA for 1bb1c0d

Commits on Jun 7, 2017

• Support unknown emit shapes in tf.nn.raw_rnn.

  Show description for 85e8322
  

  rryan

  authored and

  tensorflower-gardener

  committed
  85e8322

  Copy full SHA for 85e8322

• Support numpy-style padding and slicing of tf.spectral.rfft/irfft to match the desired FFT length.

  Show description for 7b4c017
  

  rryan

  authored and

  tensorflower-gardener

  committed
  7b4c017

  Copy full SHA for 7b4c017

Commits on Apr 17, 2017

• Enable reduce_mean, reduce_sum and reduce_prod for complex types on CPU/GPU.

  Show description for 88a6cde
  

  rryan

  authored and

  tensorflower-gardener

  committed
  88a6cde

  Copy full SHA for 88a6cde

Commits on Apr 15, 2017

• Update Eigen version to f3a22f35b044 to include updates to std::complex<T> CUDA reduction support.

  Show description for 2e89e03
  

  rryan

  authored and

  tensorflower-gardener

  committed
  2e89e03

  Copy full SHA for 2e89e03

Commits on Apr 12, 2017

• Enable tf.equal and tf.not_equal GPU kernels for complex types.

  Show description for f2f9cdc
  

  rryan

  authored and

  tensorflower-gardener

  committed
  f2f9cdc

  Copy full SHA for f2f9cdc

• Add complex128 GPU registrations to tf.gather and tf.strided_slice.

  Show description for acae9a9
  

  rryan

  authored and

  tensorflower-gardener

  committed
  acae9a9

  Copy full SHA for acae9a9

Pagination

Previous

Next